CROSS-CULTURAL ADAPTATION OF THE QUALITY OF LIFE ASSESSMENT SCALE ON ALZHEIMER DISEASE

Transcription

1 Arq Neuropsiquiatr 2005;63(2-A): CROSS-CULTURAL ADAPTATION OF THE QUALITY OF LIFE ASSESSMENT SCALE ON ALZHEIMER DISEASE Marcia Maria Pires Camargo Novelli 1, Heloisa Helena Dal Rovere 2, Ricardo Nitrini 3, Paulo Caramelli 4 ABSTRACT - Objective: To present the internal validation of the quality of life (QOL) evaluation scale for patients with Alzheimer s disease (AD) and their caregivers/family members, proposed by Logsdon et al. Method: The scale was adapted through translation, back translation and equivalence evaluation. The Portuguese version was administered to a sample of 40 patients with mild to moderate AD according to NINCDS ADRDA and DSM-III-R criteria, and also to their respective caregivers/family members. Results: The reliability of the instrument was excellent, both in the intra and the inter-examiner test-retest. The correlation coefficients for the intra-examiner assessment were 0.87/0.95/0.95 (p<0.001)) for the patient, family and caregiver versions, respectively. The correlation coefficients for the inter-examiner assessment were 0.76/0.96/0.93 (p<0.001) for the patients, family and caregiver versions, respectively. The reliability was excellent for both the patient and the relatives versions in relation to the patient s QOL (alpha=0.81 and 0.85, respectively), and also in relation to the caregiver s QOL (alpha=0.84). Conclusion: The adapted version of the instrument proved to be straightforward and brief, presenting excellent stability and reliability. The Portuguese version can be obtained by contacting the first author. KEY WORDS: quality of life, Alzheimer disease, caregiver and cross-cultural adaptation. Adaptação transcultural da escala de avaliação de qualidade de vida na doença de Alzheimer RESUMO - Objetivo: Apresentar os dados de validação interna da escala de qualidade de vida (QV) para pacientes com doença de Alzheimer (DA) e seus respectivos cuidadores/familiares, proposta por Logsdon e col. Método: A escala foi adaptada seguindo metodologia que envolveu a tradução, retrotradução e avaliações de equivalência. A versão em português foi ministrada a 40 pacientes com DA provável, segundo os critérios do NINCDS ADRDA, e de intensidade leve a moderada, segundo os critérios do DSM-III-R e a seus respectivos cuidadores/familiares. Resultados: A estabilidade do instrumento foi excelente no teste-reteste intra e inter- e x a m i n a d o r. Os índices de correlação encontrados na avaliação intra-examinador foram 0,87/0,95/0,95 (p<0,001) para as versões do paciente, do familiar e do cuidador, respectivamente. Na avaliação inter-examinador os índices de correlação foram 0,76/0,96/0,93 (p<0,001). A confiabilidade foi excelente para as versões do paciente e do familiar em relação à QV do paciente (alfa=0,81 e 0,85, respectivamente) e com relação a QV do cuidador (alfa=0,84). Conclusão: O instrumento mostrou-se de fácil e rápida aplicação, apresentando excelente estabilidade e confiabilidade após sua adaptação. A versão em português pode ser obtida com a primeira autora. PALAVRAS-CHAVE: qualidade de vida, doença de Alzheimer, cuidadores, adaptação transcultural. A l z h e i m e r s disease (AD) is a degenerative brain disease characterized by the progressive loss of memory and other cognitive functions which cause significant impairment in the patient s daily life activities and social and occupational performance as well in their quality of life 1, 2. Quality of life (QOL) is a concept that has received several definitions, depending both on the context in which it is used and the examiner s guidance 3-6. The assessment is used to evaluate the satisfaction level concerning the treatment as well as the attention received by the patient and the caregiver/family which is equivalent to a measure of the satisfaction assessment as a consumer 6. Besides its increasing use as a measure of results, the QOL assessment has been also used in decisions on allocation of resources in the Grupo de Neurologia Cognitiva e do Comportamento da Divisão de Clínica Neurológica do Hospital das Clínicas (HC) da Faculdade de Medicina da Universidade de São Paulo (FMUSP) São Paulo SP, Brasil. 1 Terapeuta Ocupacional, Mestre em Ciências e Pós-graduanda na área de Neurologia da FMUSP; 2 Assistente Social, Divisão de Clínica Neurológica do HC da FMUSP; 3 Professor Associado, Departamento de Neurologia da FMUSP; 4 Doutor em Medicina, Médico Assistente da Divisão de Clínica Neurológica do HC/FMUSP. Received 12 April 2004, received in final form 30 September Accepted 23 November Dr. Paulo Caramelli - Rua Itapeva 518/ São Paulo SP - Brasil. caramelp@usp.br

2 202 Arq Neuropsiquiatr 2005;63(2-A) public health area 7. The QOL concept is multi-factorial having several definitions in the literature 8, 9. In the dementia context, the QOL definition as proposed by Birren and Deutchman 3 is described as multidimensional, including issues such as social, environmental, health, emotional state and spiritual aspects (religiosity). Whitehouse and Rabins 3 add work, leisure activities or hobbies, cognitive skills, economic success, and psychological welfare to this definition, and finalize by stressing that QOL is determined by the individual s perception of the relative importance of each of these factors. Logsdon et al. 4 developed a QOL assessment scale for patients with AD which comprises the patient s and his/her family s version of the patient s QOL, in addition to the caregiver s version of his/her own QOL. This instrument proved to be stable having high inter and intra-examiner reliability. The purpose of this study is to present cross-cultural adaptation data, and internal validation of the QOL assessment scale on AD, as proposed by Logsdon et al. 4. METHOD The scale was proposed by Logsdon to minimize the effect of cognitive loss resulting from the disease, and was designed using plain language and simple and straightforward answers 4. The instrument consists of 13 items quantified on a four-score scale, with score 1 classified as poor, and score 4 as excellent where total score ranges from 13 to 52. Scores can be calculated for the patient s, and the caregiver s reports, and these reports can be combined into a total score incorporating the pat i e n t s and the caregiver s / r e l a t i v e s version 4, 5. As the assessment focuses on the patient, the total scoring weight is calculated by multiplying the patient report score by 2, adding the caregiver s / f a m i l y s report score, and then dividing the result by 3. Complementary to this instrument is a version where the evaluation focus is on the QOL caregiver. The originally proposed instrument is in the English language. Defails are in novelli 22. In brief, initially, the instrument as proposed by Logsdon et al. 4 underwent cross-cultural adaptation comprising the following phases: 1. Translation of the instrument, originally in English, into Portuguese by two independent translators. Then, comparison of the similarities was made between the two resulting translations and, following this, the initial version in Portuguese was produced. 2. Back-translation, whereby the initial Portuguese version was retranslated into English by two independent translators, thus revealing discrepancies in the English-to-Portuguese translation. 3. Assessment of the idiomatic, semantic, conceptual and cultural equivalences conducted by a body of judges who formed an interdisciplinary assessment committee. The committee was made up of five judges as follows: three physicians (a neurologist, a psychiatrist and a rheumatologist), a psychologist and an occupational therapist. All the judges had experience in research activities on areas related to the study. 4. Production of the Portuguese version incorporating the adjustments proposed by the judges, which had resulted from issues on which there was disagreement. Items for which an agreement index 80% was not achieved, were considered items in disagreement. 5. Pre-test of the instrument to verify the instrument understanding by the target population. The scale was administered to 32 elderly people, as follows: 12 without cognitive impairment, 10 with mild dementia, and to their respective caregivers/family. 6. Production of the final version of the QOL-AD translated and adapted. The first property that was verified was the testretest reliability of the adapted instrument. The objective was to verify the instrument stability over time. This property was evaluated by means of test-retest intra and inter-examiner administration by 2 independent examiners (MMPCN an HHDR) with a two-week interval, where this was administered to a sample of 17 patients in the intra-examiner test-retest, and to 13 patients in the inter-examiner test-retest. For the verification of the internal consistency of the adapted instrument, it was administered to a random sample of 40 patients with probable AD according to the NINCDS ADRDA13 criteria, with mild to moderate sympt o m a t o l o g y, according to both the DSM-III-R14 criteria and MMSE scores , and also according to their respective caregivers/ family at the Behavioral and Cognitive Neurology Outpatient Clinic of the Hospital das Clínicas of the University of São Paulo School of Medicine (HCF- M U S P ). The patients responded to the evaluation in the form of an interview, whereas the caregivers responded in the form of a questionnaire. The study was approved by the Ethics in Research Committee of the HCFMUSP and all participants signed the informed consent term. The data were input to Excel worksheets and analyzed by the SPSS program, version The following tests were used in the statistical analysis: 1. Kendall s agreement coefficient, in the cross-cultural evaluation of the idiomatic, semantic, cultural and conceptual equivalence of the QOL-AD. 2. Pearson s correlation coefficient and Kappa s statistical test, in the test-retest analysis. 3. In the internal consistency of the adapted instrument, the coefficient used was the Cronbach s Alfa, correlating each of the items with the total score, and again with the item life as a whole. 4. Descriptive analyses were made (means and standard deviation) of the socio-demographic variables and of the descriptive data from instrument application to the sample. 5. The significance level was set for values of p The scale can be obtained from the first author.

3 Arq Neuropsiquiatr 2005;63(2-A) 203 RESULTS Initially the data on the cross-cultural adaptation of the instrument will be presented, and subsequently the results of the application of the sample evaluation. In Table 1 we show the agreement of mean percentages among the judges in the semantic, idiomatic, conceptual and cultural evaluation of the original version, relative to the QOL-AD translated version. We observed that, in the evaluation of conceptual equivalence, the value of K was 0.316, not thereby presenting a statistically significant agreement. We verified the changes proposed by the judgeswithregardtotheadaptationofthestatements for our cultural and conceptual patterns. where Some items were seen to present disagreement and, for which, the suggestions made by the judges were accepted. The modified items were: Concerning the item energy, we observed 40% disagreement and the suggestions of the judges linked the modification of the term to willingness, so as to better represent our cultural and conceptual patterns. The item living situation also obtained the same level of disagreement as that just described; the proposed modification was the alteration of the term to dwelling. The item you as a whole obtained an index of disagreement of 60% and the suggested modification was a change to you in general. Another item with a disagreement of 60% was the ability to do chores according to the judges, a conceptual difference exists between ability and c a p a c i t y, and that the term that would be best suited to the focus of the evaluation is capacity to do tasks. The same reasoning was applied with the item ability to do things for fun, which obtained an index of disagreement of 60%. The same suggested proposal as in the previous item was followed for this item, hence modifying it to capacity to do leisure activities. The last modified item was life as a whole, with an index of disagreement of 40%, where the suggestion was modification to life in general. In Table 2 the coefficients of correlation of the evaluations are described, employing intra and i n t e r-examiner for each version, including one for the patient and for the family, with respect to the patient s QOL, and one for the caregiver on their own QOL. The test of Kappa assumes values from 1 to +1, where the negative values indicate agreement inferior to that expected by chance, whereas positive values, superior agreement to that expected Table 1. Agreement of means percentages between judges concerning the semantic, idiomatic, con - ceptual and cultural equivalence between the original version and the translated version of Quality of Life - AD and the coefficient of Kendall. Equivalence evaluations Agreement of means Coefficient of Kendall Semantic and idiomatic equivalence 86.36% K = 0.343** Conceptual equivalence 86.00% K=0.316 Cultural equivalence 78.46% K=0.359* p<0.05* p<0.01** Table 2. Pearson s correlation coefficients and Kappa s test values for patient, family and caregiver evaluations. Evaluations Intra-examiner (n=17) Inter-examiner (n=13) Patient (1) Pearson=0.87 Pearson=0.76 Kappa=0.710* Kappa=0.743* Caregivers (2) Pearson=0.95 Pearson=0.96 Kappa=1.000* Kappa=1.000* Caregivers (3) Pearson=0.95 Pearson=0.93 Kappa=0.912* Kappa=0.658* p<0.001*; (1) Patient reports on own QOL; (2) Caregiver reports on patient QOL; (3) Caregiver reports on own QOL.

4 204 Arq Neuropsiquiatr 2005;63(2-A) Table 3. Socio-demographic profile of patients and their respec - tive caregivers. Socio-demographic Mild Moderate characteristics dementia dementia (n=20) (n=20) Age (patients) ± ± 5.95 Educational level 4.90 ± ± 3.95 (patients) Gender (patients) Fem - 75% Fem - 70% Male - 25% Male - 30% MMSE a ± ± 3.33 Age (caregivers) ± ± Educational level 7.20 ± ± 4.28 (caregivers) Gender (caregivers) Fem - 80% Fem - 70% Male - 20% Male - 30% Relationship Wife - 30% Daughter - 40% Daughter - 5% Son - 15% Sister - 15% Sister - 10% Son - 5% Husband - 15% Husband - 5% Wife - 20% Granddaughter - 5% Niece - 5% a MMSE, Mini-Mental State Examination. by chance, at different levels, with 1 being perfect agreement 18. The patient s and their caregiver s socio-demographic characteristics are presented in Table 3. In regard to Table 3, the identified statistical significance in comparison to the MMSE was expected since the analyses used groups with different levels of severity (mild and moderate). The equivalence with sub groups was emphasized concerning the caregiver s scholarship degree. Unfortunately, the matching of the patient s groups and also its respective caregivers was not possible. Given that the scale is easy to a n s w e r, along with an adequate standardization of the questions and with an acceptable average scholarship for the group of mild patients, we believe that the occurrence of bias in interpretation is unlikely. The mean scores in the evaluation of QOL, and the mean time of application for the mild and moderated AD cases, are presented in Table 4. Despite the fact that the table presents a reduction in the mean score of QOL-AD in all versions, with increased severity of the disease, this difference is not statistically significant probably due to the great variability of the data. Table 4. Mean and standard deviations of the scores distributed across the 3 versions of the QOL-AD, and the duration of appli - cation (in minutes) between the groups of mild and moderate dementia. Application of QOL- AD Mild dementia Moderate dementia Patient (1) ± ± 5.59 Caregivers (2) ± ± 5.78 Caregivers (3) ± ± 6.59 Time of application (1) 5.75 ± ± 1.76 Time of application (2) 5.25 ± ± 2.73 Time of application (3) 4.50 ± ± 1.77 (1) Patient reports on own QOL; (2) Caregiver reports on patients QOL; (3) Caregiver reports on own QOL. The mean time of application of the evaluation was six minutes for the patient version, five minutes for the version of the caregiver/family, and five minutes for the caregiver version. The alpha results were satisfactory in all versions. In the patient s version we obtained an alpha=0.81, whilst for the relative s version in relation to the patient s QOL we obtained a coefficient of alpha = 0.85, and in the caregiver s version on his/her own QOL, the alpha obtained was 0.84 In Table 5 we present the correlation coeff i c i e n t s for each item in the patient s report and for the caregiver/family. DISCUSSION The cross-cultural adaptation of an evaluation instrument goes beyond mere translation by taking into consideration cultural and linguistic aspects for the target population The methodology employed in the present study followed the proposals made by Guillemin et al. 11, where the stages are characterized by the cross-cultural adaptation of QOL evaluations. With the presented data, we verified that the QOL-AD scale is a promising measure for investigation of this variable in the disease. The instrument allows evaluation of QOL using the patient s and the caregiver/family s view, thereby making the assessment more comprehensive and corresponds to two out of three possible ways of evaluating the quality of life phenomenon. This study provides reliability and stability of information in the patient s versions, and caregiver/family s version, in the patient s QOL and the evaluation of the caregiver s QOL. The QOL-AD was shown to be brief and easy to administer in patients with mild to moderate

5 Arq Neuropsiquiatr 2005;63(2-A) 205 Table 5. Correlation coefficients between the items, and the internal consistency, of the QOL-AD. Item (1) (2) (3) (Total) (Life as (Total) (Life as (Total) (Life as Whole) Whole) Whole) 1. Physical health 0.77** 0.67** 0.39* ** 0.37* 2. Energy 0.65** 0.33* 0.59** ** 0.45** 3. Mood 0.51** ** 0.47** 0.42** 0.45** 4. Living situation 0.52** 0.45** 0.77** 0.50** 0.59** 0.51** 5. Memory 0.50** 0.46** 0.70** 0.60** 0.57** 0.40* 6. Family 0.42** ** 0.69** 0.69** Marriage 0.48** ** ** 0.32* 8. Friends * 0.52** ** You as a whole 0.65** 0.46** 0.63** 0.48** 0.76** 0.66** 10. Ability to do chores 0.53** ** 0.48** 0.54** Ability to do things for fun 0.62** 0.41** 0.60** 0.37* 0.52** 0.40* 12. Money 0.51** 0.34* 0.54** ** 0.48** 13. Life as a whole 0.73** ** ** - Alpha Coefficient *P<0.05 **P<0.01; (1) Patient reports on own QOL; (2) Caregiver reports on patient QOL; (3) Caregiver reports on own QOL. Spearman correlations are given for items; Pearson correlation is given for the total score. AD. No patient or caregiver refused to complete the evaluation or demonstrated difficulties in understanding the instrument. The stability and the reliability of the instrument were very good. The intra and inter-examiner correlation coefficients, after a two-week interval, were 0.87/0.95/0.95 (p<0.001), and 0.76/0.96/0.93 (p<0.001), respectively, for the patient, the family and the caregiver versions. Comparing the results of the adapted version with those of the original instrument, we found similar values for stability of the instrument, with an intra-class of the original instrument of 0.76 for the patient, and 0.92 for the caregiver/family. The reliability of the evaluation after the crosscultural adaptation was excellent for the patient s version, as well as for the caregiver/family s version, in relation to the patient s QOL (alpha = 0.81 and 0.85, respectively) and to the caregiver s QOL (alpha = 0.84). Similarity was also identified in reliability on comparison of the results of the original and adapted versions of the instrument with the patient s versions and of the caregiver/family on the patient s QOL (alpha of 0.88 and 0.87, respectively). The disagreement among the patient s reports and caregiver/family s reports found in this study has also been observed by others 4, 19, 20, but the correlation in the total score was adequate. The disagreement among the reports could originate from the different methods of administration of the evaluation 4. We also identified that the instrument presents differences of score means in all versions, although not to a statistically significant degree, with increased dementia severity, a somewhat expected finding given the progressive nature of the disease. The importance of evaluating QOL in AD is based on the unquestionable impact of the disease on the patients and their family and caregivers daily lives. In addition, there is a need to evaluate to what extent the therapeutic interventions improve the QOL of the patient 3,4, and also their impact on the caregiver 21. With regard to limitations of the study, we found difficulties related to the sample size, which make the generalization of the data and the matching of the patient groups difficult. Studies with a larger sample should be conducted in the future.

6 206 Arq Neuropsiquiatr 2005;63(2-A) In conclusion, the Portuguese version of the scale proposed by Logsdon et al. 4 seems to be a promising measure for the investigation of QOL in AD in our country. The Portuguese version can be obtained by contacting the first author of the study. The adapted instrument was shown to be brief and easy to administer, while being stable and reliable with internal consistency of its items adequately representing the phenomenon that they intend to evaluate 22. Future studies should investigate the reasons for the disagreement found between the patients and the caregivers/family s report, and should determine the point at which the patient s cognitive impairment affects the reliability of the evaluation, and finally, which symptoms have a larger impact on the evaluation of QOL, so that future efforts can be focused on these areas. Acknowledgements - We thank Rebecca Logsdon for the authorization to use the scale in Brazil and for the constant assistance in the study development. We thank Drs. Cassio Machado de Campos Bottino, Marina Rodrigues Quaresma, Maria Auxiliadora Cursino Ferrari and Elisabeth Barham, that together with one of the authors (PC) comprised the committee that evaluated the equivalence of the Brazilian version. We thank those patients and their families who participated in the study, besides the colleagues of the Behavioral and cognitive Neurology Unit who referred patients for the research. REFERENCES 1. Herrera E Jr. Caramelli P, Silveira ASB, Nitrini R. Epidemiological survey of dementiain community-dwelling Brazilian population. A l z h e i m e r Dis Assoc Disord 2002;16: Bottino CMC, Almeida OP. Demências: quadro clínico e critérios diagnósticos. In Almeida OP, Nitrini R (eds). Demências. São Paulo: Fundo Editorial Byk, 1995: Logsdon RG, Albert SM. Assessing quality of life in Alzheimer s disease: conceptual and methodological issues. In assessing quality of life in Alzheimer s disease. New York: Springer, 2000:ix-xi. 4. Logsdon RG, Gibbons LE, Mc Curry SM, Teri L. Quality of life in A l z h e i m e r s disease: patient and caregiver reports. J Mental HealthA g i n g 1999;5: Logsdon RG, Gibbons LE, Mc Curry SM, Teri L. Assessing quality of life in older adults with cognitive impairment. Psychosom Med 2002;64: Whitehouse PJ. Quality of life: future directions. In Assessing quality of life in Alzheimer s disease. New York: Springer, 2000: Kimura M. Tradução para o português e validação do Quality of life Index de Ferrans e Powers. Tese de Livre Docência. Escola de Enfermagem, Universidade Federal de São Paulo São Paulo, Whitehouse PJ, Rabins PV. Quality of life and dementia. Alzheimer Dis Assoc Disord 1992;6: Walker MD, Salek S. Areview of quality of Life in A l z h e i m e r s disease. Part 1 issues in assessing disease impact. Pharmacoeconomics 1998;14: Ciconelli RM. Tradução para o português e validação do questionário genérico de avaliação de qualidade de vida Medical Outcomes Study 36- item Short-Form Health Survey (SF - 36). Tese, Escola Paulista de Medicina, Universidade Federal de São Paulo, São Paulo, Brasil, Guillemin F, Bombardier C, Beaton D. Cross-cultural adaptation of health-related quality of life measures: literature review and proposed guidelines. J Clin Epidemiol 1993;46: Guillemin F. Measuring health status across cultures. Rheum Eur 1995;2: Mckhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of A l z h e i m e r s disease: report of the NINCDS-ADR- D Awork group under the auspice of Department of Health and Human Services Task Force on A l z h e i m e r s disease. Neurology 1984;34: American Psychiatric Association. Diagnostic and Statistical Manual of mental disorders. 3Ed. Ve r. Washington: American Psychiatric Association, Folstein SE, Mchugh PR. Mini mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12: Bertolucci PH, Brucki SM, Campacci SR, Juliano Y. The Mini-Mental State examination in a general population: impact of educational status. Arq Neuropsiquiatr 1994;52: B rucki SMD, Nitrini R, Bertolucci PHF, Caramelli P, Okamoto IH. Normas sugeridas para o uso do Mini-Exame do Estado Mental (MEM) em nosso meio. Arq Neuropsiquiatr 2002;60: B rucki SMD, Nitrini R, Bertolucci PHF, Caramelli P, Okamoto IH. Sugestões para o uso do Mini-Exame do Estado Mental no Brasil. Arq Neuropsiquiatr 2003;61: Duarte YAO. Família: rede de suporte ou fator estre s s o r. Aótica de idosos e cuidadores familiares. Tese, Escola de Enfermagem, Universidade de São Paulo São Paulo, Kiyak HA, Teri L, Borson S. Physical and functional health assessment in normal aging in Alzheimer s disease: self-report vs. family reports. Gerontologist 1994;34: Novelli MMPC, Nitrini R, Bahia VS, Caramelli P. Cuidadores de pacientes com demência: perfil sócio-demográfico e impacto diário.a rq Neuropsiquiatr 2001;59:(Suppl 3): Novelli MMPC. Adaptação transcultural da escala de avaliação de qualidade de vida na doença de A l z h e i m e r. Dissertação, Universidade de São Paulo. Faculdade de Medicina, São Paulo, 2003.